# PneuGen框架：LLM智能体驱动软体气动机器人协同设计

> 原文链接：https://robotics.hellobike.com/papers/2026-11-11-2611-13344-6615dc4f
> 频道：papers | 更新日期：2026-11-11 | 标签：auto, arxiv
> 核心摘要：提出PneuGen智能体协同设计框架，利用LLM代理搜索软体机器人设计空间。通过设计-评审双智能体与模块响应模型，在抓取及形状匹配任务中显著优于传统优化基线，并完成真机验证，大幅降低仿真计算负载。

# PneuGen框架：LLM智能体驱动软体气动机器人协同设计

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[PneuGen: Task-Driven Agentic Co-Design of Soft Pneumatic Robots](https://arxiv.org/abs/2611.13344)（arXiv 2611.13344）

## 核心内容

针对软体机器人设计空间复杂且仿真计算负载高的问题，研究团队提出PneuGen任务驱动型智能体协同设计框架。该方法利用大语言模型（LLM）代理高效搜索设计空间，在基础形状与复杂物体的抓取任务，以及多向平面和3D形状匹配任务中，性能显著优于现有计算优化基线，并成功在真实物理环境中验证了智能体生成的软体机器人设计方案。

系统架构包含设计师智能体与评审智能体：前者基于参数化组件库生成设计程序，后者分析仿真诊断数据并提出针对性修改建议。框架引入拟合模块响应模型作为可复用物理先验，建立几何结构、气压与形变间的映射关系。设计程序经自动验证后转换为有限元仿真进行迭代评估，消融实验证实了评审智能体与响应模型在提升设计效率方面的关键作用。

## 关键看点

- **双智能体架构**：设计师Agent基于参数化组件库生成代码，评审Agent解析仿真诊断反馈修订建议，实现闭环迭代优化
- **物理先验加速**：构建模块响应模型拟合几何-压力-形变关系，为LLM提供可复用物理约束，缓解纯仿真搜索的高算力瓶颈
- **自动化仿真流**：设计程序自动校验并转换为有限元仿真，支持多向3D变形与复杂物体抓取任务的端到端评估
- **真机实测验证**：智能体生成的软体气动机器人完成实物形状匹配与抓取测试，证明算法方案具备物理可实现性

## 工程与产业观察

该工作展示了LLM Agent结合物理先验模型解决软体机器人高维设计难题的可行路径。对于具身智能硬件研发，这种“代码化设计+仿真反馈”范式不仅适用于气动结构，也为刚性本体与执行器的自动化形态优化提供了参考，有望缩短非标机器人从概念到实物的研发周期。

## 英文摘要 (Abstract)

Task-driven design of soft robots requires the consideration of factors such as geometry, morphology, material properties, and actuation, resulting in a large and complex design space that is challenging to explore. Computational optimisation methods leverage signals from simulation to iteratively improve designs to achieve task objectives. However, these methods typically demand high computational load for both optimisation and simulation. We present PneuGen, an agentic co-design framework for soft robots that leverages Large Language Model (LLM) agents to search over the soft robot design space efficiently. A designer agent proposes design programs from a parametric component library, and a separate critic agent analyses simulation diagnostics and recommends targeted revisions. Fitted module response models provide reusable physical priors relating geometry and pressure to deformation. Design programs are automatically validated and converted into finite-element simulations, enabling iterative evaluation. Across grasping tasks on primitive shapes and challenging objects, as well as shape-matching tasks that require multi-directional planar and 3D deformation, the proposed framework significantly outperforms prior computational optimisation baselines. Ablation experiments also highlight the utility of the critic agent and the module response model in facilitating efficient design. We additionally demonstrate the agent-generated soft robot designs in real-world shape-matching and grasping tasks.

作者：Shaohong Zhong, Marco Pontin, Perla Maiolino
:::

## 原始参考来源 (Sources)

- [arXiv 2611.13344：PneuGen: Task-Driven Agentic Co-Design of Soft Pneumatic Robots](https://arxiv.org/abs/2611.13344)